Structure, Function and Evolution of a Giant Enzyme, Tripeptidyl-Peptidase II

http://uu.diva-portal.org

The author (B. Tomkinson) has received permission to publish this version.

Citation for the published Book Chapter:

Eklund, S. and Tomkinson, B.

“Structure, Function and Evolution of a Giant Enzyme, Tripeptidyl- Peptidase II”

In: Isama Chiba & Takao Kamio (eds.)

Serine proteases: mechanism, structure and evolution.

Nova Science Publishers, Inc. 2012, pp. 55-70.

ISBN: 978-1-61942-669-6

URL: http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-197693

Access to the published version may require subscription.

Editors: I. Chiba and T. Kamio © 2012 Nova Science Publishers, Inc.

Chapter III

S T RU C T UR E , F UN C T I O N

A ND E V O L U T I O N O F A G I A N T E N Z Y M E , T RIPEPT ID Y L -P EPT ID ASE II

Sandra Eklund and Birgitta Tomkinson

Department of Medical Biochemistry and Microbiology, Uppsala University, Uppsala, Sweden

A

Tripeptidyl-peptidase II (TPP II) is a giant exopeptidase with an active site of the subtilisin-type. Its main function is to remove tripeptides from a free N-terminal end of longer peptides. TPP II is active at neutral pH and is dependent on the same catalytic triad as other subtilases, i.e. Asp-44, His-264 and Ser-449 (numbering for murine TPP II).

Furthermore, Glu-331 has been shown to be important for binding the N-terminal amino group of the substrate. Besides its exopeptidase activity, TPP II also appears to have a low endopeptidase activity. The large subunit (138 kDa in humans) forms a >4 MDa. Oligomerisation is essential for full enzymatic activity. The recently determined hybrid structure of the TPP II spindle from Drosophila melanogaster demonstrated that the active site is localized inside the spindle and that it is a self-compartmentalizing enzyme. TPP II is present in most eukaryotes, but has not been detected in archea and the homologous genes that appear in prokaryotes are suggested to be the result of a horizontal gene transfer. A role for TPP II in degradation of the neuropeptide cholecystokinin has been suggested, and the enzyme appears to be involved in trimming of some substrates for antigen presentation.

However, considering its widespread distribution, this is probably not its main physiological function. A more reasonable assumption is that the enzyme has evolved to participate in a general protein turnover in the cytosol of most cells, presumably together with the proteasome and other peptidases.

K eywords: Abbreviations: EM, electron microscopy; FRET, fluorescence resonance energy transfer; ROS, reactive oxygen species; TPP II, tripeptidyl-peptidase II (the

The license for this PDF is unlimited except that no part of this digital document may be reproduced, stored in a retrieval system or transmitted commercially in any form or by any means. The publisher has taken reasonable care in the preparation of this digital document, but makes no expressed or implied warranty of any kind and assumes no responsibility for any errors or omissions. No liability is assumed for incidental or consequential damages in connection with or arising out of information contained herein. This digital document is sold with the clear understanding that the publisher is not engaged in rendering legal, medical or any other professional services.

species is indicated by a lower case letter, i.e. mTPP II for the murine enzyme, hTPP II for human TPP II and dTPP II for the enzyme from Drosophila melanogaster)

I

It is now 30 years since this gigantic enzyme was discovered in rat liver. When searching for an enzyme responsible for the rapid hydrolysis of phosphorylated proteins, a hexapeptide corresponding to the phosphorylated site of pyruvate kinase, Arg-Arg-Ala-Ser(³²P)-Val-Ala, was used as a probe. The enzyme that cleaved the Ala-Ser(³²P)-bond was purified and characterized (1). It turned out to be a peptidase that removes tripeptides from a free N- terminus, irrespective of phosphorylation status, and it was given the systematic name tripeptidyl-peptidase II (TPP II¹). Despite the serendipitous discovery, the many interesting and unique features of TPP II have warranted its continued characterization.

A

M

The most prominent function of TPP II is to sequentially remove tripeptides from the N- terminus of longer peptides at neutral pH. Peptides with 5-41 amino acids can be cleaved (2,3), and the requirement for an unblocked N-terminus of both substrates and peptide inhibitors was noted early (1,2). The rate of cleavage varies 100-fold between different substrates, and is dependent not only on the sequence of the tripeptides to be removed, but also on the length of the substrate. This was shown when the N-terminal Tyr-Gly-Gly tripeptide was released faster from Dynorphin (1-8) than from Dyn A (1-5) or Dyn A (1-6) (4). There is a preference for cleaving after hydrophobic amino acids (2), and with the exopeptidase activity TPP II cannot cleave either before or after a proline residue (2).

However, in addition to the dominating exopeptidase activity, TPP II also has a low endopeptidase activity. This was first demonstrated by Geier et al. (3) and later it was shown that a specific epitope for antigen presentation was formed as a result of the endopeptidase activity (5). The endopeptidase activity of TPP II has been proposed to cleave preferentially after positively charged amino acids and can also hydrolyse prolyl-bonds (3,5). However, we have investigated the endopeptidase specificity using different peptides. When identifying the hydrolysis products through mass spectrometry, there were no clear patterns and no significant preference for cleaving after positively charged amino acids could be discerned, suggesting that the endopeptidase activity is more promiscuous than previously reported (6).

Moreover, when measuring the catalytic efficiency of the endopeptidase activity using a FRET-substrate it was found to be 4-5 orders of magnitudes lower than for the exopeptidase activity (6). Thus, the rate of hydrolysis of the endopeptidase activity is so low that its physiological significance can be questioned.

TPP II is a serine peptidase of the subtilisin-type (2,7) with Asp-44, His-264 and Ser-449, together with Asn-362, (numbering for murine TPP II, mTPP II) being essential for catalytic activity (8). The mechanism for the exopeptidase activity, i.e. why TPP II removes tripeptides and not longer or shorter peptides, has been determined. This ability has been shown to be reliant on at least one glutamate residue in the S3 part of the active site, Glu-331 (mTPP II

numbering) (9), which interacts with the positively charged N-terminus. Mutation of Glu-331 into glutamine in the murine enzyme causes a more than hundredfold decrease in catalytic efficiency, primarily as a result of an increase in KM. The corresponding alteration of the nearby residue Glu-305 results in a more than 10⁴-fold decrease in catalytic efficiency (9). In TPP II from Drosophila melanogaster (dTPP II), both mutation of Glu-312 and Glu-343 (corresponding to Glu-305 and Glu-331 in mTPP II, respectively) have a similar effect, with a 10³-10⁵ fold decrease in catalytic efficiency (10). However, the effect on KM was more pronounced in the E343Q variant.

T

S

TPP II

The catalytic domain of TPP II constitutes only a minor part of the protein. Much of the rest of the peptide chain, 1249 amino acid residues in mTPP II, is involved in the oligomerisation.

a) b)

c) d)

Figure 1. The structure of TPP II at different scales and from different perspectives. A, The 40-

of dTPP II as determined by EM (EMDataBank ID: EMD-1732 (10,67)). The repetitive units are dimers, and the size of the spindle is approximately 60×28 nm (18). B, Crystal structure of the monomer of dTPP II (10) (PDB ID: 3LXU), with the following colour scheme: catalytic domain, yellow; DH-insert, orange; central domain, green; helix domain, blue. The residues of the catalytic triad are displayed in red. C, Crystal structure of the active site

triads displayed in ball-and stick. The arrow indicates the beta-sheet hairpin in subtilisin that corresponds to region of dTPP II (yellow). The arrow indicates the same hairpin as in C. Figures B-D were prepared using Chimera (68).

This results in the formation of a giant oligomeric complex with a Mr exceeding 4 10⁶, making this one of the largest known molecular machines (Figure 1A). Because of its size, TPP II has not been easy to study. The first structure was determined by EM and revealed a double-bow or spindle-like structure of approximately 50 nm built up from repetitive units (11). However, it was not until 23 years later that a crystal structure of the dTPP II dimer could be solved. Together with a high-resolution cryo-EM map a hybrid structure of the entire dTPP II spindle was created (10).

Primary Structure

The peptide chain forming the TPP II monomer has several features that distinguishes it from bacterial subtilisin. Even though it lacks the prepropeptide of subtilisin, the length ranges between 900 and 1400 amino acid residues, dependent on species (12), compared to esidues consists of a ~200 amino acid residues insert, called the DH-insert, which is merged into the subtilisin-like catalytic domain between the catalytic aspartate and histidine and a C-terminal extension of at least 500 amino acids. Neither of these additions display sequence homology to other proteins in the current databases, and are not expected to be involved in any catalysis (12). Instead, it appears likely that they are involved in the formation of the quaternary structure (10,13).

There is one characterized splicing variant with 13 extra amino acids in the C-terminal extension of mammalian TPP II. The extra amino acids affect the complex formation, resulting in an even larger structure in addition to the oligomeric complex of normal size (14).

Secondary and Tertiary Structure

The TPP II peptide chain is folded into a >130 kDa monomer. Given the length of the peptide chain it is somewhat surprising that TPP II appears to be able to fold unassisted, even when expressed in a foreign host such as Escherichia coli (15). However, once denatured by e.g. chaotropic salts, it does not refold at a measurable rate (Eklund and Tomkinson, unpublished observation). This suggests a need for non-specific chaperones, or a sequential folding mechanism where the N-terminal domains form before the synthesis of the entire peptide chain is complete. It is interesting to note that subtilisin is dependent on its propeptide for folding (16), while TPP II apparently is able to fold without this.

Following the elucidation of the crystal structure of the dTPP II dimer, it was apparent that the DH-insert formed a separate domain and the C-terminal extension two domains: one central domain rich in beta-sheets and one C-terminal 12-helix bundle formed together with part of the DH-insert (10) (Figure 1

site inside the bowl (Figure 1B). However, monomers are rarely encountered (15,17), and appear to be unstable and less active than dimers. Instead, monomers quickly assemble into dimers in a head-to-head fashion (18).

Quaternary Structure

TPP II is dependent on complex formation for full activity (11,15,17). Dimers form extensive interactions between the central domains of the two monomers as well as part of the C-terminal helix domain. To form the oligomeric complex, dimers stack on top of one another into long strands (Fig 1A). Extensive contacts are made between dimers in the strand, involving all domains of the protein. Before the release of the hybrid structure, it was noted that mutation of Gly-252 (mTPP II numbering) into arginine resulted in disruption of complex formation (13). Nevertheless, complex formation might proceed if the dimer concentration is sufficiently high (15). Gly-252 is positioned in the DH-insert, but surprisingly, there are no direct interactions between this residue and the neighbouring dimer in the hybrid structure (10). However, it is possible to envisage that such a bulky residue as arginine might come in contact with the neighbouring dimer, and thus disturb complex formation.

The enzymatic activity is 8% of full activity in dimers, increases to 67% in hexamers and approaches 100% at a mass corresponding to 8-10 dimers (15). Clearly, the oligomerisation results in activation, but the exact mechanism has not yet been determined. In the crystallized dimers, the catalytic serine residue is displaced as a result of partial unfolding of an -helix and binding of the nearby L2 loop to the active site (10) (Figure 1C). This is probably why the crystallized dimers are inactive. It is, however, unclear whether this is the conformation of dimers in solution, since these retain approximately 10% of the full activity (15,17). It has been proposed that the L2 loop and the -helix containing the catalytic serine are very flexible in the dimer, with 90% of the dimers in a conformation unfavourable for catalysis.

When dimers associate, the L2 loop interacts with a loop in the central domain of the adjoining dimer, and the catalytic serine is fixed in a position suitable for catalysis (10).

However, this hypothesis remains to be tested to exclude the possibility that the displaced L2 loop is an artefact of the crystallization process.

The active site is secluded in the oligomeric complex, and is only accessible via a cavity system. Thus, a potential substrate would have to traverse 120 Å through openings of 32 15 Å and 22 20 Å (10). This effectively excludes bulky substrates, such as folded proteins, from being degraded by TPP II. As no other mechanism of regulation has been reported, it has been proposed that this self-compartmentalization is a way to protect the cell from unhindered proteolysis by this particular enzyme (10,13).

-10 dimers, depending on species (Fig 1A). The onformation in part of the DH-insert, and where the flexible loop connecting the central domain and the C- terminal 12-helix bundle appears to be in a fixed position (10). The function of the strand association is unclear, since the cavity system is formed separately within each strand of dimers. It is not required for full activity, as separate strands and larger, presumably (15,18). One possibility is that it protects the enzyme itself from degradation

(18).

T

E

TPP II

As stated before, there are many differences between TPP II and other subtilisin-like serine

family, is a monomeric, secreted endopeptidase. In contrast, TPP II is a cytosolic exopeptidase, albeit with a low endopeptidase activity. The exopeptidase activity has the , always cleaving three amino acid residues from the N-terminus. It is clear that TPP II has evolved significantly from the subtilases of the last common ancestor, and in this section we will explore what alterations must have occurred through time.

that: i) is capable of degrading peptides to tripeptides, ii) has a subtilisin-like catalytic domain with a DH-insert and C- terminal extension and iii) forms a large, oligomeric complex (12). The fulfilment of the first of these criteria might be inferred from the presence of two glutamate residues, Glu-305 and - 331 in mTPP II (9), as discussed under Activity and mechanism above. While the presence of these glutamates cannot be taken as proof of an existing tripeptidyl-peptidase activity, their absence makes the existence of such an activity highly unlikely.

The fulfilment of the second criterion, the presence of a DH-insert and C-terminal extension, is much more easily identified in the amino acid sequence than the tripeptidyl- peptidase activity. However, these parts of the enzyme are much less conserved than the catalytic domain, even between the established TPP II members, i.e. enzyme from yeast, plant, insect and mammals that have been characterized in vitro (1,2,12,14,19 21). Because of this, classification of possible bacterial and archeal subtilases as TPP II homologues is difficult. One recent study used hidden Markov models to look for possible homologues, and discovered some prokaryotic sequences of potential interest (22). The authors concluded that a primitive form of TPP II, containing part of the central domain, was present in the last universal common ancestor. A phylogenetic analysis of some representative sequences from this investigation revealed an interesting trend (Figure 2). It shows that the base of the tree, t of the central, beta-sheet-rich domain in addition to the catalytic domain. After these, a branch of sequences containing a DH-insert as well as central domain makes up the rest of the tree. Full-length TPP II, containing also the C-terminal 12-helix bundle, cluster together on one sub-branch. The full- length sequences are the only ones where the glutamate residues essential for the tripeptidyl- peptidase activity are present. Both prokaryotic and eukaryotic sequences are distributed throughout the tree, indicating a broad distribution of subtilases with domains similar to those of TPP II. Curiously, the prokaryote Planctomyces brasiliensis contains a full-length TPP II.

This sequence appears to be more closely related to hTPP II than dTPP II is, and has been suggested to be the result of a horizontal gene transfer from a eukaryote to the bacteria P.

brasiliensis (22). A TPP II homologue resulting from horizontal gene transfer has also been documented in Blastopirellula marina (12), which is within the same family as P.

brasiliensis. However, in the case of B. marina, the sequence appeared to originate from a plant, whereas the P. brasiliensis sequence appears to stem from a chordate, so these could not reflect the same event. Also, Glu-331, which is essential for TPP activity, is conserved in the P. brasiliensis sequence, although this was not the case for the B. marina TPP II homologue(12). The full-length TPP II thus seems to have emerged in a primitive eukaryote.

Our third criterion for true TPP II homologues, the formation of an oligomeric complex, cannot be directly inferred from sequence data. As with the exopeptidase activity, the best we can do is to eliminate sequences that most likely do not fulfil this criterion. Because all domains of TPP II are involved in oligomerisation, any sequence obviously lacking one domain, such as the C-terminal 12-helix bundle, would not be expected to have the same quaternary structure.

Figure 2. Evolution of TPP II. Phylogenetic tree of subtilisin-like serine proteases from all domains of life.

Sequences with only part of the central domain are shown in brown, enzymes with part of the central domain and the DH-insert in orange, and full-length TPP II in red. Asterisks mark the sequences with the conserved glutamate residues important for tripeptidyl-peptidase activity. Eukaryotic sequences are underlined, archeal sequences are underlined with double lines and remaining sequences are bacterial. The analysis is based on a set of representative sequences detected as TPP II homologues in a recent work (22), and was conducted in PHYLIP 3.67 using the Protpars subprogram, on a multiple sequence analysis performed in ProbCons (69,70). Bootstrap values below 90% are shown beside branches.

Thus, only the full-length sequences (red in Figure 2) of eukaryotic TPP II homologues and the prokaryotic sequences suspected to be the result of horizontal gene transfer would be considered true TPP II homologues according to our previous definition (12).

2+ for stability; it is only marginally stable in the absence of calcium (23). In contrast, TPP II is not known to possess a Ca²⁺-binding site.

The loss of the calcium-binding site may serve as an adaptation to the cytosolic environment, where calcium levels are usually low. The loop in dTPP II corresponding to the Ca²⁺-binding 283-291, dTPP II numbering) is in a different conformation (Figure 1D). Furthermore, the beta-sheet hairpin corresponding to the unstructured L2 loop in the dTPP II dimer is close to this site (Figure 1D). As has already been discussed under Quaternary structure, L2 has been implicated in the activation of TPP II during

oligomerisation. Whether there is a role in the activation mechanism also for the 283-291 loop remains to be investigated. It appears reasonable to assume that complex formation has replaced Ca²⁺-binding as a way of activating and stabilizing the enzyme.

What Came First Exopeptidase Activity or Complex Formation?

In order to understand why TPP II is dependent on the formation of such a large oligomeric complex, it could be helpful to consider its evolution. One question regarding the evolution of TPP II that has yet to be addressed is whether the exopeptidase activity or oligomerisation arose first. Since neither a TPP II homologue that forms an oligomeric complex but does not possess tripeptidyl-peptidase activity, nor a homologue with TPP II-like activity that does not form the complex has been encountered, the order can only be guessed.

It has been suggested that the last common ancestor contained a subtilase with part of the central domain (22). However, our present work suggests that this subtilase contained a DH- insert as well (Figure 2). This protein has subsequently lost the signal peptide, acquired a full central domain and helix domain, as well as features such as complex formation and exopeptidase activity. In addition, the calcium dependence must have been lost at some point during this process. It can be speculated that the first cytosolic TPP II precursor still was dependent on calcium, thus sparing the cell from unhindered proteolysis. Complex formation might have arisen as a means of regaining structural stability, and thus activity, in the cytosolic environment without endangering the proteins of the cell. Thus, an endopeptidase could have arisen that would have been sufficiently useful to the cell in order not to be counter selected. Subsequently, this enzyme could evolve an exopeptidase activity and a more complex quaternary structure. In accordance with this model, we hypothesise that at least some of the quaternary structure of TPP II evolved before the shift in activity towards exopeptidase was introduced. It is interesting to note that functional homologues of TPP II have evolved that also have complex oligomeric structures. For example, in archea the proteasome works in tandem with a large peptidase complex called the Tricorn protease, which is a hexameric complex of 720 kDa and, like TPP II, releases tripeptides (24). Whereas it makes sense that cytosolic endopeptidases are self-compartmentalizing enzymes (25), the need for cells to protect themselves against exopeptidases is not self-evident. Nevertheless, several examples of self-compartmentalizing exopeptidases exist, such as TET (tetrahedral aminopeptidase) (26), bleomycin hydrolase (27), DppA (28) and leucine aminopeptidase (29).

Evidently there seems to be some kind of evolutionary pressure towards their formation.

T

F

TPP II

Intracellular Protein Turnover

The function of TPP II must have played an important role in the evolution of the enzyme, but the physiological function of TPP II is still not completely clear. Given the widespread distribution of TPP II, the most plausible function of TPP II is a house-keeping

role in intracellular protein turnover (30 32). All eukaryotic cells have the ubiquitin- proteasome system that selects substrates for degradation and initiates the process. However, the proteasome does not degrade its substrates completely into free amino acids, but forms peptides that vary in length from 3 to 22-mers (33). These peptides are expected to be prime candidate substrates for TPP II that would perform degradation into mainly tripeptides. These would in turn be good substrates for other exopeptidases in the cytosol and subsequently completely degraded into free amino acids. In vivo, it appears that peptides > 14 amino acids are dependent on TPP II for their degradation (34,35). There is a certain amount of redundancy in the system of intracellular protein breakdown, and a number of peptidases with overlapping specificities exist. Thimet oligopeptidase (TOP), for example, is an enzyme that has been ascribed an important function down-stream of the proteasome (36). This is supported by the fact that TPP II can be deleted from a number of organisms, such as the fission yeast Schizosachharomyces pombe (12), Arabidopsis thaliana (20) and mice (37), without an obvious phenotype. Nevertheless, the presumed housekeeping role of TPP II does not exclude that it also has more specific functions. Some examples will be given below.

Degradation of C C K

A membrane-bound form of TPP II has been demonstrated to cleave and inactivate cholecystokinin (CCK) (38). CCK is a family of neuropeptides involved in a number of different processes, mainly in regulating food-intake and mediating satiety. There appears to be a partly overlapping distribution of CCK and TPP II in brain from rat (39), human and monkey (40), which could support this potential role of TPP II. Indeed, a specific inhibitor of TPP II, butabindide, was developed and could reduce food intake in mice. The effect was mediated through the CCKA type of receptors, thus demonstrating that TPP II can inactivate CCK in vivo (38). Attempts have subsequently been made to improve this inhibitor with respect to stability and efficiency (41,42). However, to our knowledge, no publications of clinical trials using any inhibitors of TPP II for treating over-eating syndromes have yet been published. Interestingly enough, a potential role for TPP II in stimulating fat formation has been reported, although this appears to be independent on the enzymatic activity of TPP II and not related to its effect on CCK (43). Furthermore, mice with a heterozygous TPP II deletion were found to be lean, but not as a result of reduced food intake (43). However, this observation could not be confirmed by investigations on homozygous TPP II-deficient mice (37,44).

Antigen-Presentation

A number of different studies have been undertaken to elucidate the role of TPP II in antigen presentation, but the results obtained are somewhat contradictory (45 47). For example, an essential role of TPP II in trimming the N-terminus of some long antigenic precursors has been shown (34,35,48), but it is dispensable in trimming others (49,50).

Furthermore, a specific epitope of the HIV-protein Nef can be formed by TPP II in a proteasome-independent way (5), but TPP II has not been found to be involved in forming - (46). Judging from experiments with knockout

mice, it appears that the main contribution of TPP II is in destroying antigenic epitopes rather than creating them (37,51). This would also be in line with the housekeeping function of TPP II.

Mitosis and Cancer

An important role for TPP II in tumour progression has been suggested, although the mechanism behind this has not yet been elucidated. It has been found that the amount of TPP (52) and also in EL-4 cells isolated from rapidly growing tumours (53). Moreover, over-expression of TPP II has been shown to increase the rate of cell-proliferation and correlate with genetic defects such as chromosomal aberrations and an increased number of centrosomes (54). A recent finding showed that TPP II is essential for c-myc induced centriole overduplication (55), although the exact function of the enzyme in this process could not be established.

An involvement in apoptosis could explain the role of TPP II in tumour progression.

Evidently, TPP II allows cells to avoid apoptosis and proceed through activated cell-cycle check-points (56). This could be because IAPs (inhibitors of apoptosis protein) are stabilized in cells with a high expression of TPP II and low proteasome activity (53). In contrast to these findings, TPP II depletion had no effect on cell proliferation in experiments with cells overexpressing c-myc or ras, and the authors concluded that TPP II is not generally important for viability of transformed cells (57). Nevertheless, the same group noted increased apoptosis of TPP II-deficient activated CD8⁺ T cells -irradiation (58). The involvement of TPP II in apoptosis is also supported by experiments with knockout mice, where it seems that the animals die prematurely due to early onset of apoptosis in some white blood cells (44). It should be noted, however, that this phenotype was not observed in a gene-trapped model (37).

Another possible mechanism for the involvement of TPP II in tumour progression is its function in stress response, although this is quite controversial. Glas and co-workers showed that TPP II translocates into the nucleus in response -irradiation in eight out of ten cell lines investigated (59,60). This translocation was inhibited by the peptide Z-Gly-Leu-Ala and correlated to reduced p53 levels. The results were challenged by Tsurumi et al, who could not detect an important role for TPP II in p53 mediated damage response (57). However, new data demonstrated that the nuclear translocation in response to stress is ROS-dependent and mediated through the MAP-kinase pathway (61). Even though some of the discrepancies can be explained by different cell densities and different levels of ROS (32), it is evident that further experiments are needed to understand the exact function of TPP II in the stress response in cells.

Taken together, the results suggest that the importance of TPP II for cell-proliferation and apoptosis is dependent on the cellular environment. It could well be that the system for intracellular proteolysis is redundant under normal conditions and therefore not crucially dependent on TPP II. However, in situations of stress there is an increased dependence on protein breakdown and the presence of TPP II thus becomes critical. This is supported by the fact that the amount of TPP II is increased in various stress situations, for example in skeletal muscle in sepsis (62), and in EL4 cells adapted to high concentrations of proteasome inhibitors (3,63). If the amount of the proteasome is reduced, which is the case in certain cancer types e.g Burkitts lymphoma (52), the dependence on TPP II could be increased.

Although it has been suggested that TPP II could compensate for loss of proteasomal activity (3,63 65) it is clear that it cannot substitute completely for the proteasome (66). Decreased proteasomal activity could lead to a changed repertoire of substrates being degraded and possibly explain the stabilization of some proteasome substrates like IAPs (53). It is possible that the observed difference of the phenotype of knockout mice could also be a result of different levels of stress. The animals in the study by Huai et al. show signs of an inflammatory response, which could be why they are more affected than the mice in the study by Kawahara et al. (37,44). An alternative explanation to the observed differences in phenotype could be the residual TPP II activity in the gene-trapped mice (37). Future investigations will most likely shed more light on the exact physiological function of TPP II in stress situations as well as under normal conditions.

C

With new sequence data and structural information regarding TPP II, an understanding of the function and evolution of this gigantic exopeptidase has started to emerge. The evolution of TPP II is a clear example of the evolution of a new function from an ancient enzyme. It is believed that the oligomeric structure has evolved to protect cells from unlimited proteolysis, and we hypothesize that this happened before the shift to exopeptidase activity. While TPP II evolved primarily to take part in intracellular protein degradation, it can sometimes fulfil other, more specific functions in neuropeptide degradation, apoptosis and stress-response.

R

[1] Bålöw, R. M., Ragnarsson, U., Zetterqvist, Ö. Tripeptidyl aminopeptidase in the extralysosomal fraction of rat liver. J. Biol. Chem. 1983 Oct. 10;258(19):11622 8.

[2] Bålöw, R. M., Tomkinson, B., Ragnarsson, U., Zetterqvist, Ö. Purification, substrate specificity, and classification of tripeptidyl peptidase II. J. Biol. Chem. 1986 Feb.

15;261(5):2409 17.

[3] Geier, E., Pfeifer, G., Wilm, M., Lucchiari-Hartz, M., Baumeister, W., Eichmann, K. et al. A giant protease with potential to substitute for some functions of the proteasome.

Science. 1999 Feb. 12;283(5404):978 81.

[4] Nyberg, F., Bålöw, R. M., Tomkinson, B., Zetterqvist, Ö. Degradation of Leu- and Met- enkephalin and their C-terminalextensions by tripeptidyl peptidase II. I: Protides of the biological fluids. Pergamon Press; p. 193 6.

[5] Seifert, U., Marañón, C., Shmueli, A., Desoutter, J. F., Wesoloski, L., Janek, K. et al.

An essential role for tripeptidyl peptidase in the generation of an MHC class I epitope.

Nat. Immunol. 2003 Apr.;4(4):375 9.

[6] Eklund, S., Kalbacher, H., Tomkinson, B. Characterization of the endopeptidase activity of tripeptidyl-peptidase II. Manuscript in preparation. 2011;

[7] Tomkinson, B., Jonsson, A. K. Characterization of cDNA for human tripeptidyl peptidase II: the N-terminal part of the enzyme is similar to subtilisin. Biochemistry.

1991 Jan. 8;30(1):168 74.

[8] Hilbi, H., Jozsa, E., Tomkinson, B. Identification of the catalytic triad in tripeptidyl- peptidase II through site-directed mutagenesis. Biochim. Biophys. Acta. 2002 Dec.

16;1601(2):149 54.

[9] Lindås, A. C., Eriksson, S., Jozsa, E., Tomkinson, B. Investigation of a role for Glu-331 and Glu-305 in substrate binding of tripeptidyl-peptidase II. Biochim. Biophys. Acta.

2008 Dec.;1784(12):1899 907.

[10] Chuang, C. K., Rockel, B., Seyit, G., Walian, P. J., Schönegge, A. M., Peters, J. et al.

Hybrid molecular structure of the giant protease tripeptidyl peptidase II. Nat. Struct.

Mol. Biol. 2010 Aug.;17(8):990 6.

[11] Macpherson, E., Tomkinson, B., Bålöw, R. M., Höglund, S., Zetterqvist, Ö.

Supramolecular structure of tripeptidyl peptidase II from human erythrocytes as studied by electron microscopy, and its correlation to enzyme activity. Biochem. J. 1987 Nov.

15;248(1):259 63.

[12] Eriksson, S., Gutiérrez, O. A., Bjerling, P., Tomkinson, B. Development, evaluation and application of tripeptidyl-peptidase II sequence signatures. Arch. Biochem.

Biophys. 2009 Apr. 1;484(1):39 45.

[13] Tomkinson, B., Ní Laoi, B., Wellington, K. The insert within the catalytic domain of tripeptidyl-peptidase II is important for the formation of the active complex. Eur. J.

Biochem. 2002 Mar.;269(5):1438 43.

[14] Tomkinson, B., Hansen ,M., Cheung, W. F. Structure-function studies of recombinant murine tripeptidyl-peptidase II: the extra domain which is subject to alternative splicing is involved in complex formation. FEBS Lett. 1997 Apr. 1;405(3):277 80.

[15] Seyit, G., Rockel, B., Baumeister, W., Peters, J. Size matters for the tripeptidylpeptidase II complex from Drosophila: The 6-MDa spindle form stabilizes the activated state. J. Biol. Chem. 2006 Sep. 1;281(35):25723 33.

[16] Ikemura, H., Takagi, H., Inouye, M. Requirement of pro-sequence for the production of active subtilisin E in Escherichia coli. J. Biol. Chem. 1987 Jun. 5;262(16):7859 64.

[17] Tomkinson, B. Association and dissociation of the tripeptidyl-peptidase II complex as a way of regulating the enzyme activity. Arch. Biochem. Biophys. 2000 Apr.

15;376(2):275 80.

[18] Rockel, B., Peters, J., Müller, S. A., Seyit, G., Ringler, P., Hegerl, R. et al. Molecular architecture and assembly mechanism of Drosophila tripeptidyl peptidase II. Proc. Natl.

Acad. Sci. U.S.A. 2005 Jul. 19;102(29):10135 40.

[19] Osmulski, P. A., Gaczynska, M. A new large proteolytic complex distinct from the proteasome is present in the cytosol of fission yeast. Curr. Biol. 1998 Sep.

10;8(18):1023 6.

[20] Book, A. J., Yang, P., Scalf, M., Smith, L. M., Vierstra, R. D. Tripeptidyl peptidase II.

An oligomeric protease complex from Arabidopsis. Plant Physiol. 2005 Jun.;138(2):1046 57.

[21] Renn, S. C., Tomkinson, B., Taghert, P. H. Characterization and cloning of tripeptidyl peptidase II from the fruit fly, Drosophila melanogaster. J. Biol. Chem. 1998 Jul.

24;273(30):19173 82.

[22] Rockel, B., Kopec, K. O., Lupas, A. N., Baumeister, W. Structure and function of tripeptidyl peptidase II, a giant cytosolic protease. Biochim. Biophys. Acta. 2011 Jul.

13; doi 10.1016/j.bbapap.2011.07.002, in press.

[23] Bryan, P. N. Protein engineering of subtilisin. Biochim. Biophys. Acta. 2000 Dec.

29;1543(2):203 22.

[24] Brandstetter, H., Kim, J. S., Groll, M., Göttig, P., Huber, R. Structural basis for the processive protein degradation by tricorn protease. Biol. Chem. 2002 Aug.;383(7- 8):1157 65.

[25] De Mot, R., Nagy, I., Walz, J., Baumeister, W. Proteasomes and other self- compartmentalizing proteases in prokaryotes. Trends Microbiol. 1999 Feb.;7(2):88 92.

[26] Borissenko, L., Groll, M. Crystal structure of TET protease reveals complementary protein degradation pathways in prokaryotes. J. Mol. Biol. 2005 Mar. 11;346(5):1207 19.

[27] lez, F., Zheng, W., Johnston, S. A., Joshua-Tor, L. Crystal structure of human bleomycin hydrolase, a self-compartmentalizing cysteine protease.

Structure. 1999 Jun. 15;7(6):619 27.

[28] Remaut, H., Bompard-Gilles, C., Goffin, C., Frère, J. M., Van Beeumen, J. Structure of the Bacillus subtilis D-aminopeptidase DppA reveals a novel self-compartmentalizing protease. Nat. Struct. Biol. 2001 Aug.;8(8):674 8.

[29] Burley, S. K., David, P. R., Taylor, A., Lipscomb, W. N. Molecular structure of leucine aminopeptidase at 2.7-A resolution. Proc. Natl. Acad. Sci. U.S.A. 1990 Sep.;87(17):6878 82.

[30] Tomkinson, B. Tripeptidyl peptidases: enzymes that count. Trends Biochem. Sci. 1999 Sep.;24(9):355 9.

[31] Tomkinson, B., Lindås, A. C. Tripeptidyl-peptidase II: a multi-purpose peptidase. Int. J.

Biochem. Cell Biol. 2005 Oct.;37(10):1933 7.

[32] Preta, G., De Klark, R., Gavioli, R., Glas, R. The Enigma of Tripeptidyl-Peptidase II:

Dual Roles in Housekeeping and Stress. J. Oncol. 2010 Aug. 18; doi 10.1155/2010/128478

[33] Kisselev, A. F., Akopian, T. N., Woo, K. M., Goldberg, A. L. The sizes of peptides generated from protein by mammalian 26 and 20 S proteasomes. Implications for understanding the degradative mechanism and antigen presentation. J. Biol. Chem.

1999 Feb. 5;274(6):3363 71.

[34] Reits, E., Neijssen, J., Herberts, C., Benckhuijsen, W., Janssen, L., Drijfhout, J. W. et al. A major role for TPPII in trimming proteasomal degradation products for MHC class I antigen presentation. Immunity. 2004 Apr.;20(4):495 506.

[35] York, I. A., Bhutani, N., Zendzian, S., Goldberg, A. L., Rock, K. L. Tripeptidyl peptidase II is the major peptidase needed to trim long antigenic precursors, but is not required for most MHC class I antigen presentation. J. Immunol. 2006 Aug.

1;177(3):1434 43.

[36] Saric, T., Graef, C. I., Goldberg, A. L. Pathway for degradation of peptides generated by proteasomes: a key role for thimet oligopeptidase and other metallopeptidases. J.

Biol. Chem. 2004 Nov. 5;279(45):46723 32.

[37] Kawahara, M., York, I. A., Hearn, A., Farfan, D., Rock, K. L. Analysis of the role of tripeptidyl peptidase II in MHC class I antigen presentation in vivo. J. Immunol. 2009 Nov. 15;183(10):6069 77.

[38] Rose, C., Vargas, F., Facchinetti, P., Bourgeat, P., Bambal, R. B., Bishop, P. B. et al.

Characterization and inhibition of a cholecystokinin-inactivating serine peptidase.

Nature. 1996 Apr. 4;380(6573):403 9.

[39] Facchinetti, P., Rose, C., Rostaing, P., Triller, A., Schwartz, J. C. Immunolocalization of tripeptidyl peptidase II, a cholecystokinin-inactivating enzyme, in rat brain.

Neuroscience. 1999;88(4):1225 40.

[40] Radu, D., Tomkinson, B., Zachrisson, O., Weber, G., De Belleroche, J., Hirsch, S. et al.

Overlapping regional distribution of CCK and TPPII mRNAs in Cynomolgus monkey brain and correlated levels in human cerebral cortex (BA 10). Brain Res. 2006 Aug.

9;1104(1):175 82.

[41] Ganellin, C. R., Bishop, P. B., Bambal, R. B., Chan, S. M. T., Leblond, B., Moore, A.

N. J. et al. Inhibitors of tripeptidyl peptidase II. 3. Derivation of butabindide by successive structure optimizations leading to a potential general approach to designing exopeptidase inhibitors. J. Med. Chem. 2005 Nov. 17;48(23):7333 42.

[42] De Winter, H., Breslin, H., Miskowski, T., Kavash, R., Somers, M. Inhibitor-based validation of a homology model of the active-site of tripeptidyl peptidase II. J. Mol.

Graph. Model. 2005 Apr.;23(5):409 18.

[43] McKay, R. M., McKay, J. P., Suh, J. M., Avery, L., Graff, J. M. Tripeptidyl peptidase II promotes fat formation in a conserved fashion. EMBO Rep. 2007 Dec.;8(12):1183 9.

[44] Huai, J., Firat, E., Nil, A., Million, D., Gaedicke, S., Kanzler, B. et al. Activation of cellular death programs associated with immunosenescence-like phenotype in TPPII knockout mice. Proc. Natl. Acad. Sci. U.S.A. 2008 Apr. 1;105(13):5177 82.

[45] Endert, P. van. Role of tripeptidyl peptidase II in MHC class I antigen processing - the end of controversies? Eur. J. Immunol. 2008 Mar.;38(3):609 13.

[46] Wherry, E. J., Golovina, T. N., Morrison, S. E., Sinnathamby, G., McElhaugh, M. J.,

Shockey, D. C. et al. Re- -

MHC class I-restricted CD8 T cell epitope. J. Immunol. 2006 Feb. 15;176(4):2249 61.

[47] Guil, S., Rodríguez-Castro, M., Aguilar, F., Villasevil, E. M., Antón, L. C., Del Val, M.

Need for tripeptidyl-peptidase II in major histocompatibility complex class I viral antigen processing when proteasomes are detrimental. J. Biol. Chem. 2006 Dec.

29;281(52):39925 34.

[48] Lévy, F., Burri, L., Morel, S., Peitrequin, A. L., Lévy, N., Bachi, A. et al. The final N- terminal trimming of a subaminoterminal proline-containing HLA class I-restricted antigenic peptide in the cytosol is mediated by two peptidases. J. Immunol. 2002 Oct.

15;169(8):4161 71.

[49] Basler, M., Groettrup, M. No essential role for tripeptidyl peptidase II for the processing of LCMV-derived T cell epitopes. Eur. J. Immunol. 2007 Apr.;37(4):896 904.

[50] Marcilla, M., Villasevil, E. M., De Castro, J. A. L. Tripeptidyl peptidase II is dispensable for the generation of both proteasome-dependent and proteasome- independent ligands of HLA-B27 and other class I molecules. Eur. J. Immunol. 2008 Mar.;38(3):631 9.

[51] Firat, E., Huai, J., Saveanu, L., Gaedicke, S., Aichele, P., Eichmann, K. et al. Analysis of direct and cross-presentation of antigens in TPPII knockout mice. J. Immunol. 2007 Dec. 15;179(12):8137 45.

[52] Gavioli, R., Frisan, T., Vertuani, S., Bornkamm, G. W., Masucci, M. G. c-myc overexpression activates alternative pathways for intracellular proteolysis in lymphoma cells. Nat. Cell Biol. 2001 Mar.;3(3):283 8.

[53] Hong, X., Lei, L., Glas, R. Tumors acquire inhibitor of apoptosis protein (IAP)- mediated apoptosis resistance through altered specificity of cytosolic proteolysis. J.

Exp. Med. 2003 Jun. 16;197(12):1731 43.

[54] Stavropoulou, V., Xie, J., Henriksson, M., Tomkinson, B., Imreh, S., Masucci, M. G.

Mitotic infidelity and centrosome duplication errors in cells overexpressing tripeptidyl- peptidase II. Cancer Res. 2005 Feb. 15;65(4):1361 8.

[55] Duensing, S., Darr, S., Cuevas, R., Melquiot, N., Brickner, A. G., Duensing, A. et al.

Tripeptidyl Peptidase II Is Required for c-MYC-Induced Centriole Overduplication and a Novel Therapeutic Target in c-MYC-Associated Neoplasms. Genes Cancer. 2010 Sep.;1(9):883 92.

[56] Stavropoulou, V., Vasquez, V., Cereser, B., Freda, E., Masucci, M. G. TPPII promotes genetic instability by allowing the escape from apoptosis of cells with activated mitotic checkpoints. Biochem. Biophys. Res. Commun. 2006 Jul. 28;346(2):415 25.

[57] Tsurumi, C., Firat, E., Gaedicke, S., Huai, J., Mandal, P. K., Niedermann, G. Viability and DNA damage responses of TPPII-deficient Myc- and Ras-transformed fibroblasts.

Biochem. Biophys. Res. Commun. 2009 Sep. 4;386(4):563 8.

[58] Firat, E., Tsurumi, C., Gaedicke, S., Huai, J., Niedermann, G. Tripeptidyl peptidase II plays a role in the radiation response of selected primary cell types but not based on nuclear translocation and p53 stabilization. Cancer Res. 2009 Apr. 15;69(8):3325 31.

[59] Hong, X., Lei, L., Künert, B., Naredla, R., Applequist, S. E., Grandien, A. et al.

Tripeptidyl-peptidase II controls DNA damage responses and in vivo gamma- irradiation resistance of tumors. Cancer Res. 2007 Aug. 1;67(15):7165 74.

[60] Preta, G., De Klark, R., Glas, R. A role for nuclear translocation of tripeptidyl-peptidase II in reactive oxygen species-dependent DNA damage responses. Biochem. Biophys.

Res. Commun. 2009 Nov. 27;389(4):575 9.

[61] Preta, G., De Klark, R., Chakraborti, S., Glas, R. MAP kinase-signaling controls nuclear translocation of tripeptidyl-peptidase II in response to DNA damage and oxidative stress. Biochem. Biophys. Res. Commun. 2010 Aug. 27;399(3):324 30.

[62] Wray, C. J., Tomkinson, B., Robb, B. W., Hasselgren, P. O. Tripeptidyl-peptidase II expression and activity are increased in skeletal muscle during sepsis. Biochem.

Biophys. Res. Commun. 2002 Aug. 9;296(1):41 7.

[63] Glas, R., Bogyo, M., McMaster, J. S., Gaczynska, M., Ploegh, H. L. A proteolytic system that compensates for loss of proteasome function. Nature. 1998 Apr.

9;392(6676):618 22.

[64] Wang, E. W., Kessler, B. M., Borodovsky, A., Cravatt, B. F., Bogyo, M., Ploegh, H. L.

et al. Integration of the ubiquitin-proteasome pathway with a cytosolic oligopeptidase activity. Proc. Natl. Acad. Sci. U.S.A. 2000 Aug. 29;97(18):9990 5.

[65] Zhang, J., Wong, J., Gao, G., Luo, H. Tripeptidyl peptidase II serves as an alternative to impaired proteasome to maintain viral growth in the host cells. FEBS Lett. 2011 Jan.

3;585(1):261 5.

[66] Princiotta, M. F., Schubert, U., Chen, W., Bennink, J. R., Myung, J., Crews, C. M. et al.

Cells adapted to the proteasome inhibitor 4-hydroxy- 5-iodo-3-nitrophenylacetyl-Leu- Leu-leucinal-vinyl sulfone require enzymatically active proteasomes for continued survival. Proc. Natl. Acad. Sci. U.S.A. 2001 Jan. 16;98(2):513 8.

[67] Lawson, C. L., Baker, M. L., Best, C., Bi, C., Dougherty, M., Feng, P. et al.

EMDataBank.org: unified data resource for CryoEM. Nucleic Acids Res. 2011 Jan.;39(Database issue):D456 64.

[68] Pettersen, E. F., Goddard, T. D., Huang, C. C., Couch, G. S., Greenblatt, D. M., Meng, E. C. et al. UCSF Chimera--a visualization system for exploratory research and analysis. J. Comput. Chem. 2004 Oct.;25(13):1605 12.

[69] Felsenstein, J. PHYLIP - Phylogeny Inference Package. Cladistics. 5:164 6.

[70] Do, C. B., Mahabhashyam, M. S. P., Brudno, M., Batzoglou, S. ProbCons: Probabilistic consistency-based multiple sequence alignment. Genome Research. 2005 Feb.

1;15(2):330 40.